Language modelActive

Qwen 3 Coder Next

Alibaba

Released: February 4, 2026
Data date: September 8, 2026

Speed: 115.8 tokens/s
Output throughput
Price: $0.18 · $0.9
Input · Output / 1M tokens
Input: Text
Output: Text

Qwen 3 Coder Next uses an unusual sparse parameter split for coding agents. It has 80 billion total parameters but activates only 3 billion per token. The technical report describes training with verifiable coding tasks, executable environments, and feedback from those environments.

The open model combines Gated DeltaNet, Gated Attention, and a sparse mixture of experts. Its non-thinking configuration handles 262,144 context tokens. Qwen 3 Coder 480B A35B is the much larger sibling checkpoint.

Specifications and access

Specification	Value and source
Weights	Open weightsSource
Access	API, LocalSource
Parameters	80BSource
Active parameters	3BSource
Context window	262,144 tokensSource
Weights repository	Qwen/Qwen3-Coder-NextSource
License	Apache-2.0Source
Architecture	hybrid Gated DeltaNet + Gated Attention sparse MoE causal language modelSource
Input	TextSource
Output	TextSource
Reasoning mode	non-thinking onlySource
Tool calling	YesSource
Structured output	YesSource
Native context window	262,144 tokensSource
Maximum context window	262,144 tokensSource
Parameters	80 BSource
Active parameters	3 BSource
Repository created	2026-01-30 (Repository creation date is a precise publication proxy, but may differ from the public announcement date or an individual checkpoint's first availability.)Source

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

SWE-bench Verified (Show measurement, test conditions, and source)

Source value: 62.8
Score: 62.8
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

SWE-bench Multilingual (Show measurement, test conditions, and source)

Source value: 56.2
Score: 56.2
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

SWE-bench Pro (Show measurement, test conditions, and source)

Source value: 64.3
Score: 64.3
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

EvalPlus (Show measurement, test conditions, and source)

Source value: 86.56
Score: 86.56
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

MultiPL-E (Show measurement, test conditions, and source)

Source value: 88.23
Score: 88.23
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

CRUXEval (Show measurement, test conditions, and source)

Source value: 95.88
Score: 95.88
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

LiveCodeBench (Show measurement, test conditions, and source)

Source value: 58.93
Score: 58.93
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

Aider-Polyglot (Show measurement, test conditions, and source)

Source value: 66.2
Score: 66.2
Metric: percent
Unit: %
Category: coding
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

MMLU (Show measurement, test conditions, and source)

Source value: 87.73
Score: 87.73
Metric: percent
Unit: %
Category: knowledge
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

MMLU-Redux (Show measurement, test conditions, and source)

Source value: 91.18
Score: 91.18
Metric: percent
Unit: %
Category: knowledge
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

MMLU-Pro (Show measurement, test conditions, and source)

Source value: 80.52
Score: 80.52
Metric: percent
Unit: %
Category: knowledge
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

GPQA (Show measurement, test conditions, and source)

Source value: 74.49
Score: 74.49
Metric: percent
Unit: %
Category: reasoning
Direction: Higher is better
Source type: vendor-reported
Evaluator: Qwen3-Coder-Next Technical Report
Status: active
Retrieved at: 2026-09-08
Methodology: Vendor-reported result. Official report evaluation; benchmark-specific harnesses and settings as reported

Source Qwen3-Coder-Next Technical Report

Page 1 of 2

Sources and data date

Type	Evidence and data date
Speed	Median across serving providers. Output speed after the first streaming chunk, not total response time. Retrieved September 8, 2026. Source: Artificial Analysis · Artificial Analysis (retrieved September 8, 2026)Speed measurement methodology
Price	Base API prices without caching or batch discounts. Higher context tiers and other rates are listed under Costs. Source: Pricing · API pricing
Input	Documented modalities. Retrieved September 8, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Qwen3-Coder-Next model card · Editorial description reviewed October 2, 2026
Output	Documented modalities. Retrieved September 8, 2026. Source: Model documentation · Model metadata (retrieved September 8, 2026) · Qwen3-Coder-Next model card · Editorial description reviewed October 2, 2026
Research status	Research date September 8, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Additional source	Qwen3-Coder-Next Technical Report (retrieved September 8, 2026) · Qwen3-Coder-Next technical report · Editorial description reviewed October 2, 2026
Additional source	Model metadata (retrieved September 8, 2026)
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Creative Commons Attribution 4.0

Qwen 3 Coder Next

Specifications and access

Measurements without matching peer values

Our tests

Pelican on a bicycle

Popularity

Published prices

Sources and data date